Diving deep into UL It’s time to see some examples of UL, given that we’ve spent some time on SL algorithms. When to use UL There are many times when UL can be appropriate. Some very common examples include the following: The first tends to be the most common reason that data scientists choose to […]
An illustrative example – beer! – Predictions Don’t Grow on Trees, or Do They?
An illustrative example – beer! Let’s run a cluster analysis on a new dataset outlining different beers with different characteristics. We know that there are many types of beer, but I wonder if we could possibly group beers into different categories based on different quantitative features. Let’s try! Let’s import a dataset of just a […]
Classification metrics 2 – Predictions Don’t Grow on Trees, or Do They?
For example, imagine we have an email with three words: send cash now. We’ll use naïve Bayes to classify the email as either being spam or ham: We are concerned with the difference of these two numbers. We can use the following criteria to classify any single text sample: Because both equations have P (send […]